Full text of "Climatic zonation for forage crops in the Atlantic Region."

■ «|L Agriculture
I t Canada

Research Direction generale
Branch de la recherche

Contribution 1983-27E

Climatic zonation for
forage crops in the
Atlantic Region

Canada

The map on the cover has dots representing
Agriculture Canada research establishments.

Climatic zonation for
forage crops in the
Atlantic Region

A. BOOTSMA
Agrometeorology Section
Land Resource Research Institute
Ottawa, Ontario

LRRI Contribution No. 83-01

Research Branch
Agriculture Canada
1984

Copies of this publication arc available from:

A. Bootsma

Agrometeorology Section

Land Resource Research Institute

Research Branch, Agriculture Canada

Ottawa, Ontario

K1A0C6

Produced by Research Program Service

©Minister of Supply and Services Canada 1984

Egalement disponible en francais sous le titre

Zonage climatologique pour la culture de.s plantes fourrageres

dans la region de I'Atlantique

TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS iv
SUMMARY, RESUME v

1. INTRODUCTION 1

2. METHODOLOGY 3

2.1 Estimating maturity dates for first cut 3

2.2 Estimating the critical fall harvest period

for alfalfa 6

2.3 Zonation criterion for harvest frequency

and potential yield 8

2.4 Zonation criteria for drying index and

water deficits 8

2.5 Mapping procedures 12

3. RESULTS AND DISCUSSION 12

3.1 Heat unit zonation for estimating optimum first
harvest date 14

3.2 Critical fall harvest period for alfalfa 16

3.3 Zones for harvest frequency and potential

yield 19

3.4 Drying index and water deficit zonation 23

4. CONCLUSION 30

5. REFERENCES 31

6. APPENDIX 35

LIST OF TABLES

Page

1. Estimated growing degree-day requirements for
specific maturity stages in several forage crops. 4

2. Equations for calculating water deficits (WD) from
Accumulated Potential Water Loss (APWL). 12

3. Average dates of accumulated growing degree-days

(GDD) for maturity zones in Figure 5. 16

4. Critical fall harvest period for alfalfa for zones

in Figure 6. 18

5. Growing season length criterion for zones in Figure

7. 19

6. Estimated optimum harvesting frequencies of several
forage species for each zone in Figure 7. 21

7. Estimated average potential dry matter yields of
several forage species for each zone in Figure 7. 22

8. Range in drying indices for zones in Figure 8. 25

9. Range in Accumulated Potential Water Loss (APWL)

and water deficits for zones in Figure 9. 26

10. Qualitative comparison between water deficit zones,
Canada Soil Climate Map moisture classes and

moisture stress on two forage species. 29

11. List of climatic stations used in zonation study

and derived data. 38

Page

111
LIST OF FIGURES
Figure

1. Correction factor for monthly growing degree-days
above 5°C (GDD) in spring. 6

2. Correction factor for monthly growing degree-days
above 5°C (GDD) in autumn. 7

3. Relationship between Thornthwaite Water Deficit and
Accumulated Potential Water Loss at three soil

water holding capacities. 11

4. General map of the Atlantic region study area. 13

5. Maturity zones for first cut of forage crops. 15

6. Zones of critical fall harvest period for alfalfa. 17

7. Zones of growing season index for cutting
frequencies and potential yields of forage crops. 20

8. Drying index zones for hay-making. 24

9. Water deficit zones for forage crops. 27

10. Approximate location of climatic stations used in
forage crop zonation study. 37

ACKNOWLEDGEMENTS

This climatic analysis of the Atlantic region is part of an attempt to
divide the region into zones for forage crop production and management
recommendations. Considerable guidance to the work was provided by an ad
hoc committee on forage crop zonation for the Atlantic region, which
consisted of Dr. M. Suzuki, Charlottetown Research Station, Mr. J.
MacMillan, New Brunswick Department of Agriculture and Rural Development
and the author. Many helpful comments on the work were received from
members of the Atlantic Advisory Committees on Forages and on
Agrometeorology , for which the author is grateful. Special thanks are also
due to Dr. J. Dumanski, Land Resource Research Institute, Mr. W.J.
Blackburn and J. A. Dyer, both from the Crop Production Division, Regional
Development Branch and Dr. M. Suzuki for reviewing the manuscript.

The helpful advice of Mr. D. Grimmett, Charlottetown Research Station
in computer data processing techniques is acknowledged. Assistance was
also provided by data processing staff of the Agrometeorology Section, Land
Resource Research Institute. Some preliminary analyses for this project
were carried out with assistance from Mr. D.B. Walker and Mr. G.S. Rodd
while the author was employed by the P.E.I. Department of Agriculture and
Forestry. Finally, thanks are due to stenographic staff of the Land
Resource Research Institute for typing services rendered. Cartographic
work was performed by the cartography unit of the Land Resource Research
Institute.

SUMMARY

This bulletin describes the spatial variation of several derived climatic
parameters of importance to forage crop production and management in the
Atlantic Provinces. Accumulated growing degree-days above 5°C (GDD) in spring
are used to estimate important maturity stages in several forage species for
first cut. GDD's remaining in the fall are used to indicate probable
variation in the critical fall period during which alfalfa and possibly other
species susceptible to winterkill should not be harvested. The length of
season between specified GDD's accumulated in spring and remaining in fall is
used as a criterion for designating variation in suitable cutting frequencies
of forages. Simplified moisture criteria based on rainfall and estimated
potential evapotranspiration are used to describe regional differences in
hay-making conditions and in the likelihood of moisture stress during summer
months.

RfiSUMfi

Dans le present compte rendu, nous decrivons la variation en fonction de
l'espace de plusieurs parametres climatiques influant sur la production et la
regie des cultures fourrageres dans les provinces de 1' Atlantique. D'apres le
nombre total de degres-jours de croissance au-dessus de 5°C (D.J.C.) accumules
au printemps, on a estime a quel moment les cultures de plusieurs especes
fourrageres sont parvenues au stade de maturation crucial pour la premiere
recolte. Les D.J.C. qui restent en automne indiquent la variation probable de
la periode critique automnale durant laquelle la luzerne et certaines autres
especes sensibles au froid d'hiver ne devraient pas etre recoltees. La
longueur de la periode entre la date au printemps ou un nombre precis de
D.J.C. sont accumules et la date en automne ou un nombre precis de D.J.C.
restent a venir sert de critere pour determiner la variation de la frequence
appropriee des recoltes de plantes fourrageres. Pour decrire les variations
regionales des conditions de fenaison et des risques de penurie d'eau durant
l'ete, on se sert de criteres d'humidite simplifies, base sur l'abondance des
precipitations et la valeur approximative de 1 'evapotranspiration potentielle.

1. INTRODUCTION

The climatic analyses described in this bulletin were undertaken in
response to a need to divide the Atlantic region into different zones for
the purpose of making specific recommendations on forage crop production
and management. Climate has an important effect on the adaptability of
various forage species and cultivars to the region. Conditions permit a
wide variety of forage crops to be grown (Atlantic Field Crops Committee
1980). While timothy is the most widely grown forage grass, other species
such as orchardgrass and bromegrass are also being cultivated. Legumes
such as alfalfa, clover, and birdsfoot trefoil are also grown, but
successful production depends upon good winter survival. In recent years
annual ryegrass has also increased in popularity as a source of feed.
Although there are many factors which affect production and management of
forage crops, climate is one of the most important. The climate not only
affects growth patterns, persistence, quality and yield, but also
influences the response of each cultivar to different management practices.

This study focuses on four main areas in which climate has a
significant influence. These areas are (i) the time when optimum maturity
for first harvest is reached, (ii) the critical fall period when alfalfa
should not be harvested for good management, (iii) the harvest frequency
and yield of forage crops as influenced by season length and, (iv)
hay-making conditions and soil water deficits during the summer period.

(i) The time when forage crops reach specific stages of maturity is an
important factor in production of high quality forages in the
Atlantic region. Research has demonstrated that grasses must be
harvested at early heading and legumes no later than early flowering
to achieve maximum yield of digestible nutrients (Atlantic Field
Crops Committee, 1980). It has also been shown that the harvest
season for timothy can be spread out by planting cultivars that grow
and develop at different rates and in this way allow the crop to be
cut at the optimum stage of growth (Grant and Burgess, 1978). The
date that a particular cultivar will reach the required stage of
maturity will vary depending mainly upon climatic conditions in the
region where it is grown. In this study, the average dates when
cultivars of alfalfa, timothy, and red clover reach specific stages
of maturity for first cut are estimated on the basis of accumulated
growing degree-days above 5°C (GDD) in each area.

(ii) The probable variation in the critical fall period during which
alfalfa should not be harvested is also examined. Alfalfa must be
hardened during the fall period to aid survival over winter in many
parts of North America. Grazing or cutting of alfalfa should be
avoided for approximately six weeks in the fall because defoliation
during this period will interfere with the hardening process and will
increase the chance of winter injury (Fulkerson, 1974; Gottfred,
1980; Heinrichs, 1969; Woolley and Wilsie, 1961). The period
required for alfalfa to attain adequate cold resistance will vary
depending upon climatic conditions. Fulkerson (1970) found that in

southern Ontario, the harvest date which resulted in the greatest
decline in alfalfa yields coincided closely with the 25 percent risk
date of frost in each region. Management recommendations advise
farmers in Ontario not to cut alfalfa for a six week period centred
around this critical fall harvest date (Ontario Field Crops Research
Committee, 1981). Woolley and Wilsie (1961) questioned the validity
of using the first killing frost to determine the date of fall
removal, and proposed a system of cold unit accumulation based on
soil temperature to determine time required for adequate hardiness.
In the Atlantic region of Canada, it has been observed that alfalfa
harvested or grazed in the previous fall sustained more winter injury
(MacKenzie and Suzuki, 1978; Suzuki and McRae, 1979; Willis and
Suzuki, 1971) . Although proper management does not guarantee
successful winter survival, it will reduce the risk of winter
injury. Therefore, crop recommendations state that alfalfa should
not be cut or grazed between September 1 and October 15 (Atlantic
Field Crops Committee, 1980). In this study, the critical fall
period during which alfalfa should not be harvested was defined on
the basis of accumulated GDD remaining in the fall. This was
considered a suitable criterion since experience in other regions
has shown that the critical harvest period is advanced in areas which
experience an earlier decline in temperature in autumn (Fulkerson,
1970; Woolley and Wilsie, 1961).

(iii) The frequency of harvesting and the potential yield of several forage
species are related to the number of days between the time when 350
GDD are accumulated in spring and 450 GDD are remaining in fall.
This criterion is based on the assumption that temperature is a prime
limiting factor in growth in spring and fall, while during summer
months growth is mainly a function of time and less influenced by
temperature (assuming no severe moisture stress) . Potential yield in
this study is defined as the yield which can be achieved with good
management under present technology, as approximated by yields
recorded at research station trials in the region in years with
relatively good winter survival.

(iv) The Atlantic region usually receives sufficient rainfall to maintain
crop growth during summer months. Nevertheless, in some years,
moisture stress significantly retards the growth of forages (Atlantic
Field Crops Committee, 1975; Black, 1978; Calder and Nicholson,
1970) . Regrowth after defoliation during summer months is affected
most often since moisture stress is most likely during this period.
For example, dry summer weather in Prince Edward Island in 1975
prevented a second cut of timothy and reduced the yield of the second
cut of alfalfa by at least fifty percent (Atlantic Field Crops
Committee, 1975). A surplus of moisture in summer may also be
detrimental to forage production in some years or regions. Black
(1978) observed that an excessive moisture supply caused a decline in
pasture production, possibly due to waterlogging of soils and
leaching of nutrients. More importantly, excess moisture during the
haying season can result in a serious decline in hay quality by

delaying the harvest past the time when the forage crop is at its
optimum maturity. Wet weather also causes a serious decline in both
the quantity and quality of forages after cutting (Wilkinson, 1981).
The above evidence confirms the importance of moisture status during
summer months in forage zonation. In this study, spatial variations
in indices of hay-drying conditions and soil water deficits are
analyzed, based on potential evapotranspiration estimates and
rainfall for the period of June through August. Moisture deficits
are a function of climate (rainfall, evaporation), soil (water
holding capacity, drainage) and crop characteristics (transpiration,
rooting habit, drought resistance). Information is incomplete on the
relationship between all these factors in the Atlantic region and
therefore simplifying assumptions were made in the analyses.

This bulletin describes in detail the methods used to assess the
spatial distribution of climatic parameters recognized as being important
to forage production and also presents the results of the analyses. The
information will help agronomists formulate forage production and
management recommendations for the region. The results also have
significance for land evaluation and assessment of crop production
potential. Maximum benefit will be achieved if the climatic information is
integrated with data on soils in the region.

2. METHODOLOGY

2 .1 Estimating maturity dates for first cut

There is very little information available in the literature on
suitable methods for predicting stages of maturity in forage crops, even
though the influence of temperature on development of grasses and legumes
has been investigated extensively (Knight and Hollowell, 1958; Kozumplik
and Christie, 1972; Pearson and Hunt, 1972; Smith and Jewiss, 1966). Most
studies have demonstrated that, except for extremely high temperatures, the
rate of development to maturity increases with increasing temperature.
Factors such as moisture, fertility, and daylength may influence the rate
of development to some extent. However, under cool, moist conditions
typically experienced during spring in the Atlantic region, temperature is
the most important factor determining when stages of maturity will be
reached. Selirio and Brown (1979) used an accumulation of 550 ODD as an
estimate of the time when alfalfa reaches the flowering stage in southern
Ontario. Comparison of accumulated GDD with crop maturity stages recorded
in forage trials conducted by the Atlantic Advisory Committee on Forage
Crops indicated that in the Atlantic Region, the early bloom stage of
Saranac and Iroquois alfalfa in post-seeding years is reached when
approximately 450 GDD have accumulated (Bootsma, unpublished
observations) . While the early alfalfa cultivars such as Saranac exhibit
more rapid growth in spring than medium cultivars such as Iroquois, most
cultivars reach maturity at approximately the same time. The data also
indicate that Champ timothy reaches the 50 percent heading stage when

approximately 450 GDD have accumulated. Clair timothy reaches 50 percent
heading about 50 to 70 degree-days sooner than Champ, and Climax is about
50 to 70 degree-days later. Clair, Champ, and Climax are timothy cultivars
recommended for production in the region that are rated as very early,
early, and medium maturity respectively. Double cut red clover cultivars
Ottawa and Lakeland reach the early bloom stage when about 450 GDD have
accumulated.

Considerable variation in the degree-day requirement between seasons
and locations indicates a need to develop a more accurate method of
predicting maturity. For example, alfalfa grown at St. John's,
Newfoundland, tends to require fewer heat units to reach the early bloom
stage than alfalfa grown at warmer Maritime locations. In some years and
at some locations there is also more spread in maturity dates and
degree-day requirements between varieties than in other years or at other
locations. In the absence of an alternative method, the accumulated
degree-day values listed in Table 1 were used to estimate maturity stages
in timothy, alfalfa and red clover in spring. These values may need to be
adjusted in future when more data from experimental plots becomes available.

Table 1. Estimated growing degree-day requirements for specific
maturity stages in several forage crops

Accumulated
growing degree
days above
5°C (GDD)

Approximate stage of maturity in post-seeding years

alfalfa*

red
clover**

Clair

timothy

Champ

Climax

350
400
450
500

early bud - early head
late bud - 50% head early head
early bloom early bloom full head 50% head early head

full head 50% head

* average based on Saranac and Iroquois cultivars.
**average based on Lakeland and Ottawa cultivars.

Average dates when maturity stages in Table 1 are reached in the
Atlantic region were estimated by determining the average date when 350 and
450 GDD have accumulated, using the following quadratic regression
equations:

Y = 126.72 - 12.527 X + 0.3077 X (1)

1 11

Y = 140.11 - 13.201 X + 0.3354 X (2)

2 11

where Y^ and Y2 are the average dates when 350 and 450 GDD

respectively have accumulated in spring (June 1=1);
X;l is the mean air temperature for May and June (°C).

Temperature normals for the 1951-1980 period (Environment Canada, 1982)
were substituted into these equations to estimate Y^ and Y2 at 231
climatic stations in the region. Temperature normals for stations with
less than 20 years of records were previously adjusted to the 30-year
normal period by Environment Canada using standard techniques.

Equations (1) and (2) were determined by multiple linear regression
analyses using data from 68 climate stations in the Atlantic region for the
1941-1970 normal period (Environment Canada, 197D. Both equations had a
coefficient of determination (r^) of 0.984 and standard error of estimate
(s.e.) of 1 day. In developing these equations, accumulated GDD were
determined from a graph of monthly mean air temperature for each station by
summing daily GDD values. A regression equation was used to correct GDD
sums based on the graphs for months with less than 200 GDD as follows:

Y c = 19.98 + 0.904 Y G (3)

where Yq is the corrected monthly GDD summation;

Yq is the monthly GDD sum determined from the mean
temperature graph.

Equation (3) was developed using monthly GDD published by Environment
Canada (Treidl, 1978) for Yq, since these were calculated from daily
maximum and minimum air temperatures. Data from 23 station-months yielded
an r 2 value of 0.99 and an s.e. of 6.5 GDD. This method of correcting
GDD sums calculated from mean air temperatures proved to be as accurate but
simpler than Thorn's method (1966). A correction factor shown in Figure 1
was also applied for months with zero GDD but with mean temperatures above
1°C. Figure 1 was determined from comparisons of mean monthly air
temperature with GDD published by Environment Canada for station months
having zero GDD based on the graph method.

The accuracy using the graph method with corrections to determine dates
when 350 GDD and 450 GDD have accumulated was further checked by comparison
with data published by Environment Canada (Treidl, 1979) for 36 locations.
Weekly GDD summations by Environment Canada were interpolated to derive the
350 and 450 GDD dates. The results of these two methods were highly
correlated (r = 0.996) and were generally within 1 day of each other,
indicating that the graph method was sufficiently accurate for the purpose
of this study.

1 i

1 i ■

• / •

• •J

/• i

i i

-20246
MONTHLY MEAN AIR TEMPERATURE (°C)

Fig. 1 Correction factor for monthly growing degree-days above 5°C (GDD)
in spring.

2.2 Estimating the critical fall harvest period for alfalfa

The critical fall period during which alfalfa should not be harvested
was defined as the ^5-day period beginning on the date when an average of
^50 GDD were still remaining in the fall. This criterion was considered
appropriate since periods were found to coincide with the critical harvest
period presently recommended for the Atlantic region (Atlantic Field Crops
Committee, 1980) for locations where alfalfa field trials have been
regularly conducted (Charlottetown, Nappan, Truro). Furthermore, periods
thus defined were centred near the 25 percent risk date for frost in inland
areas of New Brunswick and Nova Scotia, which corresponded closely with the
relationship between the critical harvest date (midpoint of the critical
harvest period) and frost risk in Ontario (Fulkerson, 1970). Tn coastal
regions, this period was centred considerably earlier than the 25 percent
risk date due to the moderating influence of ocean waters on night-time
temperatures. The criterion may need to be modified in future when more
information becomes available on the effect of cutting management on winter
survival in the Atlantic region.

A method similar to that used for estimating ODD in spring was used to
calculate the average date when 450 GDD were remaining in the autumn.
Normal temperature data for the 1951-1980 period (Environment Canada, 1982)
were used from 232 climate stations in the following regression equation:

x 3 =

-43.3^ + 6.317 X

(4)

where Y3 is the date when 450 GDD are remaining in the fall(August 1=1);
X is the mean air temperature for September and October (°C).

Equation (4) was determined by linear regression analyses using data
from 68 stations in the region for the 1941-1970 normal period (Environment
Canada, 1971). This equation had an r^ value of 0.98 and an s.e. of 1
day. Various combinations of average temperature were tried along with
quadratic terms, but these did not significantly improve the accuracy of
the regression relationship. The 450 GDD date (Y3) was determined from
normal temperature graphs by accumulating daily GDD values backwards in
time from the date when the mean temperature curve dropped below 5°C.
Similar methods were employed to correct GDD summations as with the spring
data. The regression equation used to correct GDD sums for months with
less than 220 GDD was:

Y c = 24.03 + 0.894 X G

(5)

where Yq is the corrected monthly GDD summation;

Y Q is the monthly GDD determined from the mean temperature graph.

1 ■

1 '

•

•
•

•

/ •

—J —

-—0

. 1. « .

• y

-4-2 2 4

MONTHLY MEAN AIR TEMPERATURE (°C)

Fig. 2 Correction factor for monthly growing degree-days above 5°C (GDD)
in autumn.

Equation (5) was determined by using monthly GDD summations published
by Environment Canada (Treidl, 1978) for Yq, similar to equation (3) for
the spring. Data from 26 station-months yielded an r 2 value of 0. Q Q and
a s.e. of 0.2 GDD. Figure 2 shows the correction factor used for months
with zero GDD based on the mean temperature graphs, but which have mean
temperatures above 0°C. This figure is comparable to Figure 1 for the
spring data.

The accuracy with which the graph method estimated the date when 450
GDD remain in fall was evaluated by comparison with dates determined from
interpolating weekly GDD summations published by Environment Canada
(Treidl, 1979) for 36 stations. Dates determined from Environment Canada
data were assumed to be correct values since these GDD summations were
based on daily maximum and minimum air temperatures. The graph method with
corrections applied yielded dates which were generally 1 to 3 days earlier
than dates based on Environment Canada data. The following regression
equation describes the relationship between dates determined by these two
methods:

Y A = 2.96 + 0.954 X G (6)

where Y^ is the date when 450 GDD are remaining in fall based on
Environment Canada data;
X is the date determined using normal temperature graphs with
corrections applied.

Equation (6) had an r 2 of 0.97 and a s.e. of 1.3 days. Estimates of
the 450 GDD date in the fall for all 232 stations in the region were
adjusted by using the value of Y? in equation (4) for X G in equation
(0). These adjusted values were taken as the starting date of the 4^- day
period during which alfalfa should not be harvested.

2.3 Zonation criterion for harvest frequency and potential yield

Sections 2.1 and 2.2 describe procedures used to estimate when 3 q GDD
have accumulated in spring and when 450 GDD are remaining in fall in an
average year. The number of days between these two dates (DAY?) was
calculated and used as a zonation criterion for the frequency of harvest
and potential yield of various forage species. The rationale for using
this criterion was partly discussed in the introduction. In each
particular region, perennial forage crops have available for growth the
number of days given by this parameter plus an additional 800 GDD (in
spring and fall) . For alfalfa, the parameter DAYS represents the time
available for regrowth for early cultivars after the first cut until the
critical fall period when harvesting should be avoided.

2.4 Zonation criteria for drying index and water deficits

The relative potential for curing hay in the field in various parts of
the Atlantic region was assessed by using a drying index developed from

field experimental data by Hayhoe and Jackson (1974). The index was
defined as

I = PE - 0.2 P (7)

where PE is the potential evaporation (ram);
P is the precipitation (mm) ;
I is the index value.

Equation (7) can be used to obtain daily values of I or an accumulated
value over a given number of days. For the purpose of this study, the
accumulated value from June through August was calculated for over 230
stations using temperature and precipitation normals for the 19 51-1980
normal period (Environment Canada, 1982). Total PE was estimated using an
equation developed by Baier and Robertson (1965) and the conversion factor
determined by Baier (19 71) as follows.

The average daily latent evaporation (LE) in cnP for the period June
to August was calculated by the equation,

LE = -57.334 + 1.6704 TMAX + 1.6794 TRANGE + 0.0486 Q (8)

where TMAX is the average daily maximum temperature from June through

August (°C);
TRANGE is the average difference in °C between the daily maximum and

daily minimum temperature for the same period;
Q Q is the average total solar radiation at the top of the

atmosphere over the same period (cal cm~2 day - -'-)

Q Q was estimated from station latitude (LAT) using:

Qo = 934.4 + 3-6308 (52.0 - LAT) ' 927 (9)

for latitudes between 43°N and 51.5°N, and

Q = 907.3 + 4.1759 (59.0 - LAT) ' 970 (10)

for latitudes above 51.5°N and less than 59°N.

In the development of Equations (9) and (10), Q was calculated at
specific latitudes using daily values of solar radiation at the top of the
atmosphere determined using the method described by Robertson and Russelo
(1968). Both equations had an r 2 value exceeding 0.99.

LE values from equation (#) were converted to total PE in mm from June
through August by the formula

Total PE = LE x 0.086 x N (11)

where N is the total number of days in the period.

For convenience, the drying index I was normalized to a maximum of 100
by the following formula :

I N = I_x 100 (12)

where Ijj is the normalized index;

I is the original drying index calculated from Equation (7); and
I-MAX i- s the maximum value of the drying index, taken as 360 mm.

Water deficit zonation was accomplished by using modifications of
water balance procedures described in detail by Thornthwaite (1948) and
Thornthwaite and Mather (1957). Thornthwaite defined moisture deficit as
the amount by which potential exceeds actual evapotranspiration in any
month. Calculations are time consuming and require knowledge of the soil
water holding capacity (WHC).

In this study, water deficits were calculated for three WHC's (50, 100
and 200 mm) using relationships between seasonal water deficits and
Accumulated Potential Water Loss (APWL) shown in Figure 3. Thornthwaite
and Mather (1957) defined APWL as the sum of all negative monthly values
of P-PE. In this study, APWL is defined as the sum of all positive values
of PE-P for convenience. The relationships in Figure 3 were based on
Thornthwaite water balance tabulations for selected stations in the
Atlantic provinces and Ontario by Phillips (1976). However, Phillips only
calculated water deficits for 100 and 200 mm WHC. The graph for 50 mm WHC
was constructed by calculating water deficits following Thornthwaite 's
procedures (Thornthwaite and Mather, 1957).

APWL's were calculated for 232 climate stations in the region using
total precipitation and PE values for the months of June through August
for the 1951-1980 normal period (Environment Canada, 1982). This
procedure was valid since in most cases, PE-P was positive or near zero
for these months only. PE was estimated by the same procedure as
previously described for the drying index, rather than by the Thornthwaite
method. Thornthwaite 1 s procedure overestimated PE in coastal areas of the
Atlantic region because it is based on mean temperature and does not
account for differences in relative humidity or vapour pressure deficit.
The formula of Baier and Robertson (1965) accounts for humidity
differences to some extent, since coastal areas with mean temperatures
comparable to inland locations have lower PE values due to smaller
day /night temperature ranges. PE estimates using the Baier and Robertson
formula were as much as 20 per cent higher than Thornthwaite PE for some
stations. For this reason, data from selected stations in Ontario
(Phillips, 1976) were used to determine the relationship between APWL and
water deficits in Figure 3 at these higher PE values.

Water deficits were calculated using equations fitted to the graphs in
Figure 3. Best results were achieved by using two equations for each of
the curves as shown in Table 2.

WATER DEFICIT (mm)

Table 2. Equations for calculating water deficits (WD) from Accumulated
Potential Water Loss (APWL)

WHC x APWL d

(mm) range Equation used to estimate water deficits

(mm)

50 0-70 WD= 0.404 + 0.11351 AWPL + 0.005220 (APWL) 2

50 >70 WD= -16.658 + 0.6520 APWL + 0.0010685 (APWL) 2

100 0-150 WD = 0.01697 ( APWL) 1 ' 67581

100 >150 WD = 0.85338 APWL - 52.776

200 0-150 WD = 0.0061224 (APWL) 1,78276

200 >150 WD = 0.61737 APWL - 46.222

1 WHC - water holding capacity of the soil.

2 APWL - accumulated potential water loss.
NOTE: For APWL ^0.0, WD = 0.0.

2.5 Mapping procedures

Criteria previously described were determined for approximately 232
climatic stations in the Atlantic region. Results were plotted on maps of
the region and isolines connecting points of equal value were drawn. Major
topographic features such as the Annapolis and Saint John River valleys
were considered when drawing isolines. Data from climate stations in
Quebec and Maine were used to help position isolines of the criteria
described in sections 2.1, 2.2 and 2.3 in the vicinity of the borders.
Adjustments were also made to account for biases in these same criteria at
first order hourly synoptic stations in the region. These biases are due
to different observational procedures with respect to the climatological
day for minimum temperature at synoptic and ordinary climate stations
(Bootsma, 1976). For example, a bias in the monthly mean air temperature
at stations with hourly observations could delay estimated spring maturity
dates (section 2.1) by as much as 5 days. Where discrepancies were
evident, more weight was given to ordinary climate stations than to first
order stations in the mapping procedure.

3. RESULTS AND DISCUSSION

A general map of the Atlantic region which identifies the locations
referred to in the text is shown in Figure 4. The province of Newfoundland
was mapped at a considerably smaller scale than the other three Maritime
provinces. A list of climatic stations used in this study and calculated
values for designated criteria are given in the Appendix. The approximate
locations of the climatic stations are identified in the Appendix in Figure
10.

3.1 Heat unit zonation for estimating optimum first harvest date

Since isoline patterns for the dates when 350 GDD and 450 GDD have
accumulated in spring as determined from equations (1) and (2) were
similar, one zonation map could be drawn for the region (Figure 5). The
450 GDD date was generally 10 days later than the 350 GDD date. The
average dates when 350, MOO and 450 GDD have accumulated in each zone in
Figure 5 are given in Table 3. The 400 GDD dates were derived by
interpolation.

The date when 350 GDD have accumulated in spring varied from before
June 15 in the Annapolis and Saint John River valleys to as late as the
last week in July in some parts of Newfoundland. The 350 GDD date
corresponds to the date when Saranac and Iroquois alfalfa are estimated to
reach the early bud stage in an average year in each region. This
represents the earliest possible date for the first cut if a maximum period
of regrowth is desired for additional harvests later in the season.
Alfalfa cut on this date would have higher digestibility and protein
content than if cut at 450 GDD, but lower dry matter yields (Atlantic Field
Crops Committee, 1980). Clair timothy is also expected to be at the
desirable maturity stage for first cut on this date.

The date when 450 GDD have accumulated in spring varies from before
June 25 in the Annapolis and Saint John River valleys, to as late as the
first week in August in some areas of Newfoundland. This date corresponds
approximately to the optimum date of first cutting of Climax timothy and
double cut red clover cultivars, and the latest desirable cutting date of
alfalfa.

The data presented are based on expected heat unit accumulations in a
normal year. Variation in maturity will occur from season to season
depending on weather conditions. Within each zone there will also be
variability in heat units available to the crop due to differences in soil
and in microclimate. For example, forages grown on cool, wet soils which
are slow to warm up in spring will mature later than those grown on warmer
soils although yields may be higher on the former. Field exposure will
also affect the accumulation of heat units, and good shelter due to
windbreaks and/or topography can advance maturity considerably. Weather
stations are usually located on fairly well-exposed sites and the zonation
map should be quite representative of this type of condition. The map
should be helpful in formulating recommendations on suitable harvesting
dates for the first cut of forage crops in each area of the Atlantic
region. However, growers will need to base their decision on when to
actually take the first cut on the observed stage of maturity as it is
influenced by local environmental conditions and management.

Table 3. Average dates of accumulated growing degree-days (GDD) for
maturity zones in Figure 5

Zone

Average da'

tes when GDD have

accumulated

35C

) GDD* 1

400 GDD* 2

450 GDD*3

June

15 or

earlier

June 21 or earl:

ier

June 25 or earlier

June

16-20

June 22-26

June 2 6-30

June

21-25

June 27-Jul 1

July 1-5

June

26-30

July 2-6

July 6-10

July

1-5

July 7-11

July 11-15

July

6-15

July 12-21

July 16-2 5

July

16-25

July 22-31

July 2 6- Aug 4

July

26 or

later

Aug 1 or later

Aug 5 or later

Estimated average date when alfalfa in early bud and Clair timothy
in early head stage.

Estimated average date when alfalfa in late bud stage, Clair
timothy in 50% head and Champ timothy in early head stage.
Estimated average date when alfalfa and double cut red clover in
early bloom, Champ timothy in 50$ head and Climax timothy in early
head stage.

Note: Alfalfa refers to the cultivars Saranac and Iroquois;

red clover refers to the cultivars Lakeland and Ottawa.

3.2 Critical fall harvest period for alfalfa

The average date when 450 GDD remain in the fall as determined by
equations (4) and (6) was used to construct the zonation map shown in
Figure 6. Since the 450 GDD date was taken as the beginning of a 45-day
period during which alfalfa should not be harvested, it was possible to
define a critical fall harvest period for each zone as shown in Table 4.
The estimated starting date of the critical period ranged from as early as
August 5 in northwestern New Brunswick to as late as September 5 in some of
the coastal regions in southwestern Nova Scotia. In most of the
agricultural areas within the region, the estimated starting dates ranged
between August 21 and September 4 (zones F5 to F7). In Labrador, the date
when 450 GDD remain in fall ranged from as early as July 10 in the north to
after July 25 in the southeast.

, 5

-1

" Q

-n i

s^E

• 1

<fc(a

\ ss

3* «

"ST] >

ui 1 §

LU
0-

~\v£s

\l^

j^p Jj"

\_j£>

V — K™ \T

\ u_

CNJ^

\1j

u. Yy/ \ M

LL.

-«

]V\\

-s

I
I

g[fe

1 \ N l

1 V

111

O
CO

fc-

S 2 I

/)P

\ in\ \ >L-

-3

S^:

^jW^ <-

^^L

&~v

/
?„ /

\ p*

^>v

1 -*\

-s

52xj

^-^ xn

X 3\ /

C$fe

1 ^ V

^Si^xii^-

\g \ /

CQ
W
D

n ^

( I

7 1/

1
-1

»-

v \«0-

(J/

/*~A\~

-8

^^ Li. /

K\M

JSf^^l^i

(Ss\ (

y"\

The zonation map has been prepared on a relatively broad scale.
Variations in the critical harvest period within each zone are likely,
depending upon local conditions of shelter and frost. Seasonal weather
patterns can also cause some shifts in the critical period from year to
year within a particular zone, although such variation is not likely to be
large (Fulkerson, 1970).

Table 4. Critical fall harvest period for alfalfa for zones in Figure 6

Zone Critical Fall Harvest Period

Start of period 1 End of period * 2

Fl Aug. 5 or earlier Sept. 20 or earlier

F2 Aug. 6-10 Sept. 21-25

F3 Aug. 11-15 Sept. 26-30

F4 Aug. 16-20 Oct. 1-5

F5 Aug. 21-25 Oct. 6-10

F6 Aug. 26-30 Oct. 11-15

F7 Aug. 31-Sept. 4 Oct. 16-20

F8 Sept. 5 or later Oct. 21 or later

Based on date when 450 GDD are remaining in the fall.
Based on 45 days after the starting date.

In areas of the region where alfalfa production is feasible, it is best
not to harvest alfalfa during the critical period indicated by the zonation
map as this prevents the plants from reaching adequate hardiness before
entering the dormant stage (Fulkerson, 1970). Harvesting around the middle
of the critical period results in the most damage since this will cause
plants to enter dormancy with the lowest levels of food reserves
(Fulkerson, 1974). Harvesting before the critical period allows the plants
to enter later growth stages during which it develops buds on basal stems
and crowns and begins to store sufficient food reserves in the roots.
These buds must be well protected during winter, as the number of survived
buds determines the number of new shoots in the spring. Added protection
can be provided to the buds by leaving the fall growth to act as a
windbreak and catch the snow over the winter. If winterkill of alfalfa is
not a problem, the critical harvest period could possibly be shortened
somewhat at either end without causing a significant decline in survival.
However, this would have to be determined by experience within a local area.

The zonation map should be helpful in formulating recommendations on
harvest management of alfalfa in late summer and fall. Following the
suggested cutting practices does not guarantee good survival, but will
assist in reducing the risk of winter injury. The zonation map may also be
useful for other crop species that are subject to winterkill and that
require a fall rest period. For example, experimental data from Ontario
have indicated that birdsfoot trefoil also requires a fall rest period,
although the critical fall harvest date when most damage is sustained may
be more than a week earlier than for alfalfa (Fulkerson, 1982). Other
species to which this zonation may apply include clover and less hardy
grass species such as orchardgrass and perennial ryegrass.

3.3 Zones for harvest frequency and potential yield

The variation in the index of growing season length (DAYS) in the
region is demonstrated by the zonation map in Figure 7. The range in
values of this index associated with each zone is shown in Table 5. This
index exceeds 75 days in part of the Saint John River Valley, in the
Annapolis Valley, and in part of the north shore of Nova Scotia. Most of
the remainder of the major agricultural regions in the Maritimes fall in
the H2 zone which has between 60 and 74 days. In Newfoundland, the index
DAYS ranges from less than zero in the extreme north to over 45 days in the
Humber Valley region, the Grand Falls-Gander Lake area, and in the vicinity
of St. John's. In Labrador the date when 350 GDD have accumulated in
spring generally falls 10 to 30 days after the date when 450 GDD remain in
autumn.

Table 5. Growing season length criterion for zones in Figure 7

Zone

Days*

39-20

19-0

less than

* In addition to indicated number of days, there are 350 GDD
in spring and 450 GDD in fall available for crop growth.

Results from regional variety trials conducted from 1971 to 1981 by the
Advisory Committee on Forage Crops (formerly the Forage Sub-Committee of
the Atlantic Field Crops Committee) and from field experiments on forage
management by Kunelius et al. (1976, 1977b, 1978, 1980) and by MacLeod et
al. (1972) were examined to assess potential yields and optimum harvest

frequencies within specific zones in the region. Since field trials were
conducted in only three of the six zones, the proposed guidelines are
somewhat speculative and need to be verified or adjusted as more
information becomes available from field trials or farming experience.
Data on red clover were particularly scarce due to difficulties in
achieving adequate winter survival within experimental plots in many
years. Data on orchardgrass were also relatively few in comparison with
most other forage crop species.

Tables 6 and 7 list estimated optimum harvest frequencies and potential
yields, respectively, in each zone in Figure 7 for several forage species
grown in the region. Potential yields are yields which should be
achievable with good management under present technology. Yield estimates
are based on yields recorded in field trials conducted at research stations
in years with relatively good winter survival. Overwintering damage can
seriously reduce yields of alfalfa, clover and orchardgrass in some years,
thus making it difficult to achieve potential yields consistently year
after year. The proposed cutting frequencies attempt to balance quality
and yield. More frequent harvests will often improve forage quality but
result in a decline in yield or persistence.

Table 6. Estimated optimum harvesting frequencies of several
forage species for each zone in Figure 7

Forage species Optimum harvesting frequency

HI H2 H3 H4 H5 H6

Annual ryegrass * 4

Alfalfa 3

Orchardgrass 4-3

Timothy & brome grass 2

Red clover 2

4-3

3-2

2-1

3-2

1-0

4-3

2-1

1-0

■1

Cutting frequency in seeding year; all others in post-
seeding years.
Double cut cultivars such as Ottawa and Lakeland.

As indicated in Table 6, the optimum cutting frequency for annual
ryegrass in the most favourable production zones HI and H2 is four times
(Kunelius, 1980; Kunelius and Calder, 1978) but up to five cuts are
feasible. Potential yields range from about 8 tonnes dry matter per
hectare in the more favourable production areas to less than 5

Table 7. Estimated average potential dry matter yields of several
forage species for each zone in Figure 7

Forage

Average

potentia

1 dry

matter

yield (t

n. \*1

Dnnes/ha)

species

Annual ryegra

ss*2

8.0

7.0

5.0

3.0

Alfalfa

8.5

6.5

4.0

Orchard grass

7.0

7.5

6.5

5.5

4.0

3.5

Brome grass

7.5

8.0

7.0

6.0

5.0

4.0

Timothy

8.5

9.0

8.0

6.0

5.0

Red clover*3

8.0

8.5

7.5

6.0

4.0

1 Yields are based on assumption that cutting frequencies are

similar to those listed in Table 6.
^ Yields in seeding year; all others in post-seeding years.
*3 Double cut cultivars such as Ottawa and Lakeland.

tonnes per hectare in zones with shorter growing seasons. One reason why
annual ryegrass is becoming popular in the region is because of its ability
to supply quality forage for livestock in late fall when most perennial
grasses are unproductive (Kunelius, 1980; Kunelius and Calder, 1978).

The suggested cutting frequencies for alfalfa are based on the
assumption that at least 40 days are required for regrowth after the first
cut and prior to the fall rest period when harvesting must be avoided.
Thus only one cut is feasible in zone H4 while two cuts can be harvested in
zone H3. In zones Hi and H2 there is a possibility of two additional
harvests before the fall rest period. These guidelines for harvesting
frequency are general and need to be modified according to local
experience. For example, if persistence or winter survival is poor, the
number of cuts may need to be reduced, while an extra harvest may be
feasible late in the fall following the critical harvest period if
winterkill is not a problem (MacLeod et al., 1972). Fewer harvests may
give better results in years when growth is restricted by abnormally cool
temperatures or lack of moisture. In the most favourable production areas
(zones HI, H2 and H3) average yields of 8.5 tonnes dry matter per hectare
are feasible under good management, normal climate, and good winter
survival (Table 7). In zones H4 and H5 potential yields are depressed
partly because only one harvest can be taken.

Orchardgrass is a fast growing cool season perennial which can be
harvested 3 or H times per season in the most favourable areas for
production in the region (Kunelius and Suzuki, 1977b). However, it is
susceptible to winterkill and may require a long fall rest period similar
to alfalfa (Kunelius and Suzuki, 1977a). If winter survival is less than
adequate, the number of cuts may need to be reduced from those indicated in
Table 6. Average yields of 7 tonnes dry matter per hectare or more are
achievable in the more favourable production areas (zones HI and H2) while
yields are limited by cool short growing seasons in zones H5 and H6.

Timothy is the main forage species used in the region and is capable of
producing 8 tonnes dry matter per hectare or more in zones HI to H4 with
good management. Two cuts are feasible in these zones. More frequent
cutting reduces yields but produces higher quality forage (Kunelius et al. ,
1976) . Bromegrass is noted for its strong second growth and superior
performance under dry conditions in comparison to timothy, although it is
somewhat less hardy. Two cuts are possible except in the zones with the
shortest growing season (H5 and H6).

Two cuts of red clover are feasible in zones HI to H3 if early-
flowering cultivars such as Lakeland and Ottawa are grown. In zones H4 and
H5 only one cut is feasible due to shorter growing seasons. Field trial
data indicate that the second cut usually can be taken after 50 days or
more of regrowth. Red clover yields of 8.0 tonnes dry matter per hectare
can be achieved in the more favourable climatic zones. However, red clover
is a short-lived legume which is frequently winterkilled and therefore is
most useful in short-term rotations (Atlantic Field Crops Committee,
1980) . It is also more difficult to field cure as hay than most other
forage species.

The general guidelines on cutting frequencies and potential yields
apply to near average weather conditions. Abnormal weather conditions in
any given year can affect the optimum cutting frequency and yield of most
species. In most cases the final decision on cutting frequency must be
reached by considering a number of factors including quality, yield, stand
persistence, winter survival, seasonal weather conditions, and storage
method (hay or silage) .

The proposed zonation map (Figure 7) will be useful in formulating
recommendations on cutting frequencies and potential yields of forage crops
in various parts of the region. Further improvements in this zonation
scheme are required which take into greater consideration the influence of
factors such as winter survival, available moisture and forage quality on
optimum harvesting frequencies and potential yields.

3.4 Drying index and water deficit zonation

A zonation map indicating relative drying conditions for making hay in
the region is shown in Figure 8. Zone boundaries were determined by
arbitrarily selected values of the normalized drying index as indicated in
Table 8. This table also shows the range in the drying index itself and

3 s

52°

NDL

KrEy

O s

r ^

54°

EWF

0,*

^°\\$

<
2

>
<

QlO^

j u, O §

7 < S 5

V^-V ^

[in

/ O CO <^

<^C\j\(,V^l

cf\_

s 7

^^^^

-s

~^>\

Q
Z

fc ?

' I

CV|\

z
c

\ \

-2

m
Q

A ('

S 1

00 V,

2 /'cm /

CU, Q/

f U'

-as

- - " CNJ

->.\q

'-If

'\(

-■ /

— -

' Q ^

Q ,

^-/sl

\tn

-k

r;^

' k-

— s^

^*:

--'

1<M

\ \

1 \T>

-fe

v 1

PQ
W

£^

| ^_

\ V_/ s

-k

~^_y

I «

1 k "

-k

)

•

gives a qualitative rating of haymaking conditions for each zone. A more
detailed analysis of hay-drying conditions in relation to the climate in
each zone is needed for' a more quantitative assessment of hay-drying
potential. However, the present map provides a good relative indication of
drying conditions in the region.

The best drying conditions are found in zone Dl which has a normalized
index of 90, and an actual drying index value of 325 mm. This zone is
found along a large part of the Saint John River valley and stretches into
a large section of eastern New Brunswick. The zone is also present in
several regions of Nova Scotia, including part of the Annapolis Valley, an
area south of Truro and a portion of the north shore region, which
represent the main farming areas of this province. Prince Edward Island is
predominantly zone D3, with a normalized index between 65 and 80. Less
favourable drying conditions in that province are not from higher rainfall,
but because of lower PE due to depressed daytime temperatures and higher
humidities. Lower drying potential in much of Newfoundland, most of which
falls in zones D3 to D5, is similarly more due to lower PE values than to
higher rainfall.

In comparison to PE, average rainfall is relatively uniform throughout
the Atlantic region for the period June-August. Most of the region
receives an average of 230 and 280 mm of rainfall over the 3-month period,
while PE ranges from below 250 mm to over 400 mm.

Range in drying indices for zones in Figure

Map Drying Normalized
Zone Index (mm) Index*

Description of haymaking
Conditions

325

Fair-good (most favourable field
drying conditions in the region)

290-325

80-90

Fair

235-290

65-80

Fair-poor

180-235

50-65

Poor

180

Very poor

Very poor (lowest potential for
field drying in the region)

*Normalized Index = Drying Index x 100, where Max. Drying
Max. Drying Index Index is 360 mm.

The zonation map provides a good indication of the relative drying
conditions in the region and should be useful for making recommendations on
harvesting and conservation methods when combined with additional
information on hay-making conditions in each zone. Further studies to
quantify the relationship between climate and hay-drying potential in each
zone would be of benefit in this regard. Variation in drying rates for
hay-making among crop species needs to be considered in the application of
a drying zonation system.

A zonation map for water deficits is shown in Figure 9. Ranges in
water deficits and APWL values in each of the 5 zones used are shown in
Table 9. The zone boundaries were selected on the basis of 25 mm intervals
in water deficits for soils with 100 mm WHC. Water deficits for 50 mm and
200 mm WHC's are also shown in the table.

Table 9 indicates that forage crops grown in areas zoned W4 and W5 can
experience significant water deficits if soils have less than 100 mm WHC.
However, in zones Wl and W2, even shallow rooted crops like timothy are
unlikely to experience significant water stress in an average year.

Table 9.

Range in Accumulated Potential Water Loss (APWL) and
water deficits for zones in Figure 9

APWL
Water Deficit range
Zone (mm)

Approximate water deficit range (mm)

Soil water holding capacity (WHC)
50 mm 100 mm 200 mm

0-78

78-116

116-150

150

nil nil

0-40 0-25

140-75 25-50

75-105 50-75

105 75

nil
0-15
15-30
30-45
45

According to Figure 9, under similar soil conditions (ie. WHC and
drainage) forage crops are most likely to be affected by water stress in
part of the lower Saint John River valley and eastern regions of New
Brunswick, and in the Annapolis Valley and part of the north shore regions
of Nova Scotia. However, these regions are least likely to be affected by
problems relating to waterlogging of soils, leaching of nutrients and
denitrification during summer months under given soil conditions.

Areas zoned Wl or W2 are least likely to experience significant water
stress, while the probabilities of moisture surpluses which may cause
excessive leaching and denitrification are the greatest. In New Brunswick,
zones Wl and W2 are found in the extreme northwest region and along the
southern coast bordering the Bay of Fundy. In Nova Scotia, these zones are
mainly confined to areas along the southern and western coast and to areas
in Cape Breton Island. Most of Newfoundland is in zones Wl and W2, with
the exception of a few isolated regions. In zone Wl, APWL values are less
than zero, indicating that rainfall exceeds PE even during the summer
months.

It is useful to compare these estimates of Thornthwaite water deficits
with irrigation requirements or water deficits calculated by Coligado et
al., (1968) who used a daily water budgeting technique described by Baier
and Russelo (1968). Average water deficits in this study for soils with
200 mm WHC were comparable to those by Coligado et al. (1968) at a storage
capacity of 100 mm (4 inches), a consumptive use factor (CU) between 1.0
and 0.75 and a risk level of 50 percent. Water deficits for 100 mm WHC
were similar to Coligado 's at a storage capacity of 50 mm (2 inches) and a
CU factor of 0.75. Water deficits for soils with 50 mm WHC were comparable
to Coligado 's at a storage capacity of 25 mm (1 inch) and CU factor
between 0.75 and 0.5. Since storage capacity as used by Coligado et al.
(1968) is equivalent to 50 percent of WHC used in this study, the water
deficits determined by these two methods are remarkably similar.

Sly and Coligado (1974) developed a simple method for computing
seasonal water deficits which was used to prepare agroclimatic maps of
Canada. A soil climate classification system was subsequently designed for
Canada on the basis of seasonal water deficits (Baier and Mack, 1973;
Clayton et al., 1977; Mack, 1970). Water deficit zones in this study are
compared with the moisture subclasses used in the soil climate map of
Canada in Table 10, thus linking the results to a nationally accepted
classification system. This comparison provides a means of linking water
deficits determined by the Thornthwaite approach to the Canadian soil
climate classification system in the Atlantic region. A qualitative
description of the moisture stress likely experienced for timothy and
alfalfa in each zone is also presented in Table 10.

Irrigation requirements or water deficits computed by Sly and Coligado
(1974) assume that when the readily available water (50 percent of WHC)
stored in the soil is depleted, the additional water required is added by
irrigation. Since forages are produced under non-irrigated conditions in
the Atlantic region, the Thornthwaite procedure was considered more
applicable to this study. The present study uses data from more climatic
stations in the region and from the most recent 30-year normal period. In
spite of differences in methods and analyses period, the present results
appear to be relatively compatible with results from previous water deficit
zonation studies of the region.

The precise extent to which water deficits limit forage yields in each
zone is presently not known. Further research is required to help
determine these relationships and thereby assist in rating land for forage

til

■H

■P

•H

£_

4-S

•H

hO LTi

•H f-

v-»

CO w

C w CO ^

•H O
CO MD

bO O
•H cr>

1 ^

CO -_^

O CO C\J rl PO

■H QO
rH on

CO ^

bO t>0

•H -H O

^ 10 in

bO O
•H OO
CO ^

L CU -P S.
O bO CO CD

<D UJ frt
CO CO (2

CD 0)
O "O O ■<->

1 M-h X3

CO cO

03
a

=»

co G
S -H

S-

■rH

O CO

•H

■H

C 4->

T3 U L OL 3

production potential and in evaluating yield response to irrigation.
Variations in characteristics such as water use, drought resistance and
resistance to waterlogging among crop species need to be taken into
consideration in the application of water deficit zonation systems.

4. CONCLUSION

The results of climatic analyses presented here will be useful for
formulating more specific recommendations on production and management of
forage crops in the Atlantic region. However, available experience in crop
production in each part of the region will also need to be considered. The
zonation is on a scale which does not account for variations in crop
response within zones resulting from differences in micro-climatic
conditions. Variations in climate between years which influence the
production and management of forage crops was also not taken into
consideration in this study. It should be noted that the transition
between zones on the maps is usually gradual and not abrupt as may be
implied.

The criteria used and the interpretations applied will require
improvement and adjustment as additional information on forage crop
production and management becomes available from research trials and field
experience and as existing climatic resources within the region are better
defined. The results should not be interpreted as a recommendation that
the species and cultivars referred to are well adapted to all zones in the
region. Rather, the study provided guidelines on applicable management
practices if local experience has found production of specific cultivars to
be feasible.

Simple procedures for calculating climatic criteria used in zonation
are described which make use of monthly climatic normals. These facilitate
the use of data from stations with relatively short term normals which have
been adjusted to the latest 30-year normal period. The procedures also
simplify calculations for other normal periods as may be required. Several
areas requiring additional research have been identified. The zonation
maps should be helpful in selecting suitable locations for conducting field
experiments. Finally, the results of this study have significance for land
evaluation, particularly when integrated with information on soils in the
region.

5. REFERENCES

Atlantic Field Crops Committee. 1975. Report of field trials conducted
in 1975. Forage Sub-Comm. Report, p. G-6.

Atlantic Field Crops Committee. 1980. Field crop guide, Atlantic
provinces. A.F.C.C. Publ. 100, Agdex 100, pp. 33-45.

Baier, W. 1971. Evaluation of latent evaporation estimates and their

conversion to potential evaporation. Can. J. Plant Sci. 51:
255-266.

Baier, W. and Mack, A.R. 1973- Development of soil temperature and soil

water criteria for characterizing soil climates in Canada. In: Field

Soil Water Regimes, special publ. No. 5, Soil Sci. Soc. Amer. , p.
195-212.

Baier, W. and Robertson, G.W. 1965. Estimation of latent evaporation
from simple weather observations. Can. J. Plant Sci. 45:276-284.

Baier, W. and Russelo, D.A. 19 68. A computer program for estimating

risks of irrigation requirements from climatic data. Tech. Bull. No.
59, Agrometeorol. Sect., Plant Res. Inst., Can. Dept. Agric. , Ottawa,
48 pp.

Black, W.N. 1978. Effects of irrigation and nitrogen on a natural
pasture sward. Can. J. Plant Sci. 58: 347-356.

Bootsma, A. 1976. A note on minimum temperature and the climatological
day at first order stations. Atmosphere 14(1): 53-55.

Calder, F.W. and Nicholson, J.W.G. 1970. Pasture productivity of three
swards with and without nitrogen fertilizer. Can. J. Anim. Sci. 50:
467-473.

Clayton, J.S., Ehrlich, W.A., Cann, D.B., Day, J.H. and Marshall, I.B.
1977. Soils of Canada, Vol. 1: Soil Report, p. 73-77.

Coligado, M.C., Baier, W. and Sly, W.K. 1968. Risk analyses of weekly

climatic data for agricultural and irrigation planning. Tech. Bull.
Nos. 17-24, Agrometeorol. Sect., Plant Res. Inst., Can. Dep. Agric,
34 pp.

Environment Canada. 1971. Temperature and precipitation 1941-1970,

Atlantic provinces. Atmospheric Environment Service, Downsview,
Ont. , 55 pp.

Environment Canada. 1982. Canadian climate normals: temperature and

precipitation 1951-1980, Atlantic provinces. Atmospheric Environment
Service, Downsview, Ont., 136 pp.

Fulkerson, R.S. 1970. Location and fall harvest effects in Ontario on

food reserve storage in alfalfa (Medicago sativa L. ) . In:
Proceedings of the XI International Grassland Congress, Univ.
Queensland Press, p. 555-559.

Fulkerson, R.S. 1974. Stop alfalfa winterkill. Ontario Min. Agric.
Food, Factsheet #74-040, Agdex 121/21, 2 pp.

Fulkerson, R.S. 1982. Fall harvest effects on the yield and persistence
of birdsfoot trefoil ( Lotus corniculatus L. ) Forage Notes 26(1):12-13.

Gottfred, N. 1980. Winter survival of forage stands. Manitoba
Agriculture, Forage Facts, Agdex 120/21, 3 pp.

Grant, E.A. and Burgess, P.L. 1978. Timothy: High-quality forage for
livestock in Eastern Canada. Agriculture Canada publ. 1640, 15 pp.

Hayhoe, H.N. and Jackson, L.P. 1974. Weather effects on hay drying
rates. Can. J. Plant Sci. 54: 1479-484.

Heinrichs, D.H. 1969. Alfalfa in Canada. Canada Dept. Agric. Publ.
1377, 28 pp.

Knight, W.E. and Hollowell, E.A. 1958. The influence of temperature and
photoperiod on growth and flowering of crimson clover ( Trifolium
incarnatum L. ). Agron. J. 50:295-298.

Kozumplik, V. and Christie, B.R. 1972. Heading response of orchardgrass
seedlings to photoperiod and temperature. Can. J. Plant Sci. 52:
369-373.

Kunelius, H.T. 1980. Effects of nitrogen rates and harvest schedules on
yield and quality of Westerwolds ryegrass grown as a summer annual.
Can. J. Plant Sci. 60: 519-524.

Kunelius, H.T. , Suzuki, M. and Winter, K.A. 1976. Influence of harvest

systems and nitrogen rates on yields, quality, and persistence of
Champ timothy in the seeding and postseeding years. Can. J. Plant
Sci. 5b: 715-723.

Kunelius, H.T. and Suzuki, M. 1977a. Seeding year yields and quality of
orchardgrass as influenced by N rates and harvest systems. Can. J.
Plant Sci. 57: 427-431.

Kunelius, H.T. and Suzuki, M. 1977b. Response of orchardgrass to

multiple harvests and rates of nitrogen in post-seeding years. Can.
J. Plant Sci. 57: 763-770.

Kunelius, H.T. and Calder, F.W. 1978. Effects of rates of N and regrowth
intervals on yields and quality of Italian ryegrass grown as a summer
annual. Can. J. Plant Sci. 58: 691-697.

Mack, A. R. 1970. Report of the subcommittee on soil climate in relation
to soil classification and interpretation. In: Proceedings of the
eighth meeting of the Canada Soil Survey Committee, Ottawa, Ont., pp.
21-34.

MacKenzie, D.N. and Suzuki, M. 1978. Winter survival of forage legumes,

winter cereals and strawberries in the Maritime provinces in 1978.

1978 Research Summary, Agric. Can. Res. Sta. , Charlottetown, ip.
30-31.

MacLeod, L.B., Kunelius, H.T. and Calder, F.W. 1972. Effects of early

summer and fall cutting management on dry matter yields,
digestibility, crude protein, and survival of Saranac and
Narragansett alfalfas. Can. J. Plant Sci. 52: 941-948.

Ontario Field Crops Research Committee. 1981. 19 81 Field crop
recommendations. Ont. Min. Agric. Food, Publ. 296, p. 10.

Pearson, C.J. and Hunt, L.A. 1972. Effects of temperature on primary
growth of alfalfa. Can. J. Plant Sci. 52:1007-1015.

Phillips, D.W. 1976. Monthly water balance tabulations for

climatological stations in Canada. Environment Canada, Atmospheric
Environment, DS No. 4-76 (Revised).

Robertson, G.W. and Russelo, D.A. 1968. Astrometeorological estimator

for estimating time when sun is at any elevation, elapsed time
between the same elevations in the morning and afternoon, and hourly
and daily values of solar energy, Q . Tech. Bull. 14,
Agrometeorol. Sect. , Research Branch, Agriculture Canada, Ottawa, 22
pp.

Selirio, I.S. and Brown, D.M. 1979. Soil moisture-based simulation of
forage yield. Agric. Meteorol. 20:99-114.

Sly, W.K. and Coligado, M.C. 1974. Agroclimatic maps for Canada-derived
data: Moisture and temperature regimes. Tech. Bull. No. 8l,
Agrometeorol. Sect., Plant Res. Inst., Agric. Canada, Ottawa, 36 pp.

Smith, D. and Jewiss, O.R. 1966. Effects of temperature and nitrogen

supply on the growth of timothy (Phleum pratense L.). Ann. Appl.
Biol. 58: 14 5-157.

Suzuki, M. and McRae, K.B. 1979. Factors influencing winter survival of
forage legumes in the Maritime provinces. 1979 Research Summary,
Agric. Can. Res. Sta., Charlottetown, p. 32.

Thorn, H.C.S. 1966. Normal degree days above any base by the universal
truncation coefficient. Monthly Weather Review 94(7): 461-465.

Thornthwaite, C.W. 1948. An approach towards a rational classification
of climate. Geographical Review 38(1): 55-94.

Thornthwaite, C.W. and Mather, J.R. 1957. Instructions and tables for
computing potential evapotranspiration and the water balance. Drexel
Inst. Tech., Centerton, N.J., Publ. Climatol. 10(3): 181-311.

Treidl, R.A. (ed.) 1978. Handbook on agricultural and forest

meteorology, Part I. Atmospheric Environment Service, Downsview,
Ont., Table 3-

Treidl, R.A. (ed.) 1979. Handbook on agricultural and forest

meteorology, Part II. Atmospheric Environment Service, Downsview,
Ont., Table 5.

Wilkinson, J.M. 1981. Losses in the conservation and utilization of
grass and forage crops. Ann. Appl. Biol., 98: 365-375.

Willis, C.B. and Suzuki, M. 1971. Effect of cutting management on forage
yield, carbohydrate reserves and root vitality of alfalfa. 1971
Research Summary, Agric. Can. Res. Sta., Charlottetown, p. 21-22.

Woolley , D.G. and Wilsie, C.P. 1961. Cold unit accumulation and cold
hardiness of alfalfa. Crop Sci. 1: 165-167.

6. APPENDIX

Climatic data from 232 stations in the Atlantic region listed in
Table 11 were used in the forage crop zonation study. A map is supplied
in this appendix which shows the approximate location of each climate
station (Fig. 10). Parameters that were derived from station data are
listed in Table 11. Following is a brief explanation of each parameter
listed:

Column

1. Station number . This number was assigned on the basis of an
alphabetical listing of stations by province. It corresponds with
numbers in Fig. 10.

2. Station name . Station names were taken from published normals for
the 1951-1980 period by Environment Canada (1982). Where the same
name is assigned to more than one station, the station elevation is
given in brackets following the name.

3- Type of normal . The code for length of record is based on
temperature and precipitation normals published for the 1951-1980
period by Environment Canada (1982). Where the normal period for
these two variables differ, the temperature code is given first,
followed by the code for precipitation, e.g. 1/2. The codes refer to
the following periods of record:

1 complete 30 years

2 25 to 29 years

3 20 to 24 years

8 adjusted normals based on 5 to 19 years inclusive from
1951 to 1980, and any other available data from 1931
to 1950

First order synoptic stations with hourly observations are identified
with an asterisk (*) following the type of normal. These stations
may have a bias in any data derived from temperature due to a shift
in the climatological day for minimum temperature from ordinary
climate stations.

4. 450 GDP date in spring . This is the average date when 450 growing
degree-days above 5°C have accumulated in spring based on regression
equation (2) using May/ June mean air temperature for the 1951-1980
normal period. The 350 GDD date is generally about 10 days earlier.

5. 450 GDD date in fall . This is the average date when 450 growing
degree-days above 5°C are remaining in fall, as estimated using
regression equations (4) and (6) and September/October mean air
temperature for the 1951-1980 normal period.

6. Days . This represents the index of growing season length as
determined by the time interval in days between the date when 350 ODD
have accumulated in spring and when 450 GDD are remaining in the
fall.

7. Drying index . The value of the drying index, in mm, is determined
using equation (7).

8. Normalized drying index . The values in this column are calculated
using equation (12). Since it is a "normalized" index, it has no
units. In a few cases the index value may exceed 100.

9. PE. The potential evaporation in mm from June through August is
estimated by the Baier and Robertson method (equations (8) and
(ID).

10. Precip . The average total precipitation (P) from June through
August, in mm, for the 1951-1980 normal period.

11. APWL. The accumulated potential water loss in mm is determined for
the period June through August using the formula APWL = PE - P.

12. Water deficit . The average water deficits (mm) for soils with 50,
100 and 200 mm water holding capacity are calculated from APWL values
using equations shown in Table 2. Water holding capacity is here
defined as the amount of soil water between field capacity and the
permanent wilting point.

111

Ou)(f\OHjOc

O g

5<S e

)3raiflnj4>x:-Cj:n!o'<3t. t -n) t -t. < flni34) 1, -^i R| ot>cjOw-^-H_go

fi
g j

N bO

«> Sa o£ 3 o >>h >^o>>4> n£w e bdH^n

Ccr.HEm0.i->a>nJ.* . C 5 L - U |n a, 2 on,tl0 o

o3a>L r H(U(i)-Hon'-p- i:> mn!(i)033^otHQ--HO

rszoo.KKKtcnnni'i'Oi'ii'inBiHHHDjj

2 % 3 *

: O M L 4.

3 £ rn a c

ill

60 S O

c£?S § S § gr"^

.h ■* ?$ v% 2 tlranrt caS-H.S-wqo q « nj *> 2 ° Js ° S x % ^ o ° 3 <» -< •£ °c 9! £ $

' #w #3

SI
II
s|

(II

OoCOf-t-voOf

> .p oj l c a g :

30<1><U- O -^ 3 3 -r

:s: 2: z;OOTWOTOTf

£ j

2 S52

^ojicoa '-'trf

3 O f- 3- ^> ^-1 ^

r f- o O (\j O r

-H o c c o <

, $1

L "a. M c a> o « « »hhh
<o«-ioi.t.<i>o oo

:ss

t, -a -H

ocooa-inoot-

3oc\jooinoc\joooc\ja-oc

roozrocnooc

do mo oih ^o r

! SB J3 a. 3 -H ^^555°^^jS^o U C <1, > 1 n)^W'
,bO + C° e^^T) BrH-HO.CS.rjD'i

CONVERSION FACTORS

Approximate

conversion

Metric units

factors

Results in:

LINEAR

millimetre (mm)

X 0.04

inch

centimetre (cm)

X 0.39

inch

metre (m)

X 3.28

teet

kilometre (km)

X 0.62

mile

AREA

square centimetre (cm 2 )

X 0.15

square inch

square metre (m 2 )

X 1.2

square yards

square kilometre (km 2 )

X 0.39

square mile

hectare (ha)

X 2.5

acres

VOLUME

cubic centimetre (cm 3 )

X 006

cubic inch

cubic metre (m 3 )

X 35.31

cubic teet

cubic metre (m 3 )

X 1.31

cubic yards

CAPACITY

litre (L)

■ 035

cubic toot

hectolitre (hl_)

X 22

gallons

hectolitre (hL)

x 25

bushels

WEIGHT

gram (g)

* 04

oz avdp

kilogram (kg)

> 22

lb avdp

tonne (t)

■ 1 1

short tons

AGRICULTURAL

litres per hectare (L

ha)

« 089

gallons per acre

litres per hectare (L

ha)

< 357

quarts per acre

litres per hectare (L

ha)

• 071

pints per acre

milhlitres per hectare (mL ha)

• 0014

II oz per acre

tonnes per hectare (

t ha)

- 45

tons per acre

kilograms per hectare (kg ha)

• 89

lb per acre

grams per hectare (g ha)

X 0014

oz avdp per acre

plants per hectare (plants ha)

• 405

plants per acre

LIBRARY / BIBLIOTHEQUE

AGRICULTURE CANADA C

3 1073 000E83 C 10 5